Impaired socio-emotional (SE) functioning is a prominent feature of schizophrenia (SZ), especially the ability to perceive emotion in faces and prosody. Conceivably, such deficits result in difficulties integrating into society and thus maintaining friendships and relationships at work. Compared to other psychiatric population such as bipolar disorder (BP), SZ patients experience greater impairments in SE functioning and poorer functional outcome than BP patients. Therefore, understanding how deficits in SE neurocircuitry contribute to different levels of social and occupational functioning in SZ relative to BP and healthy controls (HC) can provide important knowledge for developing specific training programs and treatments that target different psychiatric populations. This project will apply multimodal neuroimaging - Magnetoencephalography (MEG) and Diffusion Tensor Imaging (DTI) - to assess the underlying mechanisms of altered SE functioning in SZ and BP patients. MEG'S unique combination of temporal and spatial resolution, in combination with structural MRl (sMRI), provides information on brain processes occurring millisecond by millisecond and is able to evaluate top-down and bottom-up emotional processing at different stages. DTI adds complementary information about how white matter integrity contributes to deficits at different stages of emotion processing. To better understand how SE impairment mediates social functioning in SZ, the proposed research will (1) evaluate the SE neural networks unique to and shared in SZ and BP by measuring brain activity and structures associated with SE processing via the use of MEG, sMRI, and DTI; (2) evaluate the social functioning differences between SZ and BP by assessing the ability to perceive emotion and the ability to integrate social skills in society using cognitive and psychosocial measures; and (3) evaluate associations between neural networks and associated anatomy and performance on psychosocial and social functioning measures. Together, the aims of this project will provide a novel dataset that will elucidate the common and unique aspects of SE functioning in SZ, BP, and HC.